Understanding drug information

How to read between the lines

HAI Europe contacts have often played an important role in critically evaluating information about medicines. Many groups have run campaigns exposing published material that misuses scientific data and turns drug information into drug promotion. Others have taken an active role in providing independent drug information for health professionals or consumers.

Health Canada, the federal health agency in Canada, is in the process of re-examining its health protection policies including those affecting drug information and drug advertising. This is partly due to strong pressure within the country to allow direct-to-consumer (DTC) advertising. Some of this pressure stems from the pharmaceutical and advertising industries and some from the "seepage" of DTC advertising that floods over its border with the United States where DTC is legal.  Like every country in Europe which bans DTC advertising, a change to such laws will put further pressure on  consumers to make drug-related decisions  on the basis of  promotional, not educational  information.  Many see this development as having an adverse impact on patient health as promotional activity tends to maximize, not rationalize drug use.

As part of a funding programme called the Health Transition Fund, set up to study new innovations in the delivery of health  care, Health Canada funded a multiple-university team to research consumers' drug information needs.  This two-year project developed  evidence-based drug information to address the needs of consumers, and then measured the impact that this drug information had on health decision-making.  The study aimed to find how consumers and health professionals are affected by access to unbiased drug information and if such access altered the treatment patients received as evidenced by changes in prescribing patterns or non-drug treatment counseling.

The Canada Drug Guide Project carried out focus group discussions with consumers, physicians and pharmacists and interviewed more than 1,400 people across Canada who used their information sheets.  They studied consumer prescription drug information needs and measured how their information sheets impacted drug decision-making.   According to HAI contact Alan Cassels, the project's manager, consumers often face an uphill battle in getting the information they need to make good decisions about their own drug therapy. Despite the fact that there is a flood of material available on medicines, much of it is either incomplete, biased, or unreadable for the patient.  Often the information is sponsored by disease groups or drug manufacturers that promote certain therapies regardless of the evidence of their effectiveness.  Consumers are often unaware of the bias which can sneak into drug information material. 

So what does good drug information look like? According to the Centre for Health Information Quality based in the UK, information should be evidence-based, involve the needs of the end-users in its production and be clearly communicated.  The use of evidence, for example, in comparing a medicine's benefits and harms should give consumers room to make informed decisions about their treatment. Below Cassels describes why some drug information available to consumers, which presents the probable benefits of a medication one way and the harms in another, might be seen as  promotion in disguise.  In any case there is reason for the consumer to be skeptical in their reading of reports of new medicines.

Confusing consumers

Some say if you want educated patients, you better educate doctors first. While this may be true, more and more people are seeking health information on their own. The increasing amount of information about drugs available on the Internet, in magazines, or on the evening news means that consumers are confronted with health claims and statistics that are sometimes difficult to place in the perspective of how it will affect their own health.

In order for someone to make an informed decision about a prescription medicine, they might want to start by understanding what the newspaper headlines mean. Many times the benefits of drugs based on clinical trials are expressed in a way that compares apples with oranges--that is, the benefits are presented as apples and the risks are given as oranges, two different things that can't really be compared well. Presenting information in different ways sometimes causes confusion as to a drug's harms and benefits. This is not very helpful to the person who has to make the decision about whether or not to take a drug for a long period of time. In order to start from the clearest possible understanding of what they are getting themselves into, people must get information that compares apples to apples and oranges to oranges.

People need information that gives context to their disease or condition.  This must be done in a way that helps people appreciate the magnitude of the potential benefits, harms and costs of drug therapy.

Information about drugs can be confusing. Test your own knowledge of how well you can decipher the benefits of the drug described below. [i] After reading each scenario below, decide if you would take the drug or not.

Imagine that your doctor says there is a drug that, if taken every day for the next five years, significantly reduced your risk of having a heart attack.  Imagine that this drug had no side effects and the best clinical studies showed that:

All of those scenarios present the same data in different ways. When you see percentages, you should immediately ask yourself:  what are the drug's  numbers in terms of absolute risk reduction (ARR),  relative risk reduction (RRR), and number-needed-to-treat (NNT).  Knowing these numbers will give you a much better appreciation of the magnitude of the benefit and of the potential for a positive impact taking a specific drug.

Scenario 1: The 34% is the relative risk reduction.  In total, 84 people (4.1%) out of 2,030 taking the placebo had heart attacks; 56 (2.7%) out of 2,051 taking the drug had heart attacks.  The RELATIVE difference between those who did and those who didn't have a heart attack is 34% (as 2.7 is 34% less than 4.1).      

Scenario 2:  1.4% is the absolute risk reduction.  When you compare the absolute percentages of  those 4.1% of placebo-takers who had heart attacks against the 2.7% of   drug-takers who had heart attacks, the absolute difference is 1.4%.

Scenario 3:  How many people need to take the drug to prevent one heart attack? You  would have to give this drug to 100 people over five years to prevent 1.4 heart attacks (4.1-2.7). Alternatively, your doctor would have to treat 71 people just like you for five years to prevent one of those people from having a heart attack. This number is called the number needed to treat (NNT).  The problem, of course, is we have NO WAY of knowing in advance who the one person is who will be prevented from having a heart attack.  It should also be noted that no drug is without side effects or costs.

To point out the misleading way in which trial data can be used, Cassels did his own research to compare what a recent newspaper article about the estrogen drug raloxifene (Evista) suggested to Canadian consumers and then examined the actual numbers behind the claims[ii].

The article suggested that raloxifene (Evista)reduces incidence of breast cancer by 76% with only a 1% risk of side effects.

What it didn’t make clear is the fact that the patients who took the drug in the study went from having a 1% absolute risk of having breast cancer down to a 0.24% absolute risk of having breast cancer over three years.  (hence, the “76%” reduction). If measured in relative terms, many of the side effects increased much more than 76%--in fact some risks, such as those for blood clots, increased, relatively speaking, by 300%. While this may not be as serious as getting breast cancer, the bottom line is that the patients taking the drug, must know in absolute terms what the benefits and harms are likely to be so that they can make an informed decision.

To understand the risks and benefits, it is helpful to use an analogy: What if you had a mortgage of US$100,000 and someone offered you a 76% reduction? You’d jump at the chance to save US$76,000. But what if someone offered you a mortgage reduction of 76% and you only owed US$1 on your mortgage. You would save 76 cents. What if you had no mortgage at all? How useful is having a 76% reduction of nothing?  The same is true with medicines: If your risk is low to start with,  a 76% reduction in something already rather low  is pretty insignificant.  It is misleading and bordering on the fradulent to promote a 76% reduction when the person barely benefits, especially when such exaggerated benefit comes with so much downplaying of the risks or possible harm.

Understanding absolute versus relative risk reduction is very important for the consumer as drugs have side effects and potential harms.  The decision to take a medication comes down to an understanding of the drug’s benefits as weighed against the harms.  Recent studies have shown that consumers will make vastly different decisions regarding long-term drug therapy depending on whether information from a trial is presented to them as relative benefits or absolute benefits. [iii]  In choosing a medication, a physician should be able to explain the benefits and harms to their patient in the same terms so they can fully participate in the decision making regarding their drug therapy.

For more information about the Canada Drug Guide project and its conclusions, visit its website: http://www.canadadrugguide.org or contact Alan Cassels at e-mail: cassels@canada.com or telephone (+1-250) 952 2951.

References